© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 10211
    Prev
    Next
    INDEX
    Explanations

    information related to academic books and publications

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    640
    +0.21
    0.8%
    393
    +0.20
    0.7%
    447
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.21
    0.13
    640
    +0.20
    0.13
    116
    +0.14
    0.06
    Negative Logits
    etooth
    -1.05
    inarily
    -0.99
    matical
    -0.90
    apsed
    -0.89
    ame
    -0.89
     heroine
    -0.86
    agara
    -0.83
    matically
    -0.83
    ensical
    -0.82
     admiration
    -0.82
    POSITIVE LOGITS
     Reloaded
    1.21
     Definitive
    1.21
     Lessons
    1.17
     Volume
    1.11
     Including
    1.08
     Principles
    1.07
     Remastered
    1.06
     Become
    1.06
     Featuring
    1.04
     Faster
    1.00
    Activations Density 1.330%

    No Known Activations